500KHz Inverter

At this frequency, a tuned resonant circuit might be a good choice if the load is fairly constant. The output can be close to perfection if the load is linear. If you use PWM (pulse width modulation), you won't need to change the DC voltage to control the output voltage. You can easily regulate with a 10:1 range. You might even do without the transformer if the load is floating but you would get a lot of leakage current in the stray capacitance if it exceeds the nF range.

The exact designs depends on the application. Provide more details such as the type of load, the power required, the regulation expected, the acceptable voltage distortion...

The design will be very different depending on the application details.

The target is to build an hyperthermia machine (see wikipedia). The load is the human skin and the voltage is applied to the skin using two electrodes operated by a doctor.
If the wave is not a perfect sinusoidal waveform, it can produce undesirables effects like spasms, a lot of heat, etc. And It´s very important that the frequency be 500KHz because the electrodes are optimized to works in this band.
The output power has to be adjustable between 10W and 100W.

Do you think that a push-pull inverter with a fixed frequency of 500KHz (by a microcontroler by example) might it work?
Javier.

study up on your radio transmission principles.

take a AC generator, setup for 500KHz (or the required frequancy) and use your 48VDC as your supply rails to the amplifyer cct, you may get some cross-over distortion, but at 500KHz, it would be hardly noticeable.

or you can run a EPROM with the voltage steps for 1 cycle, clock the address lines and punch the data lines to a A-D converter, use that as your time/frequancy base, this way you can split your +ve and -ve sides of your full bridge circuit.

It´s a possible solution. The output power has to be adjustable between 10W and 100W. In these ranges, Which will be the estimated efficiency of the system? Can you send me a proposed circuit please? The target is to build an hyperthermia machine (see wikipedia). The load is the human skin and the voltage is applied to the skin using two electrodes operated by a doctor.

If the wave is not a perfect sinusoidal waveform, it can produce undesirables effects like spasms, a lot of heat, etc. And It´s very important that the frequency be 500KHz because the electrodes are optimized to works in this band.

Thanks.

There are bipolar electrosurgical/diathermy generators in the 500 kHz range with the power you are looking for.

My advice would be:

1) start the experiments with a certified medical generator off the shelf and try to get hold of the service manual for the circuits. Alternatively research for circuits first then get the unit

2) Check for relevant patents before you copy circuitry

3) don´t forget to read the relevant standards for electrosurgery EN 60601-2-2 and ANSI/AAMI HF18-2001 early enough. They contain many hints for people new in the field that help in the design stage also for the accessories.

Good luck !

Start with a sine oscillator then use a high power amplifier with variable gain, instead of varying the power supply.

Better yet, goto a design house and get them to build the rig for you.

I'm not prepared to "Donate" circuits that will be used in medical experiments. If something goes wrong, I wouldn't be too happy, even less if I was the one being zapped.

Almost been in the state of hypothermia, my shakes fight it off now :P

Hi,

Just click on this http://www.i4at.org/lib2/inverter.htm to give you a gist. And like somebody said it before, I too disclaim the responsibility for you trying to use this circuit directly as it was meant for a different purpose.

Many years ago, when I did my basic electronics course I had to build a circuit like this, 2V p-p output in my case. The good thing about this oscillator is that as soon as the power supply voltage reaches the e-b junction voltage (.6-.7V) it comes alive, more likely than the patient you wanna use it on, providing you do your home work properly.

I do not know your level of knowledge in electronics but the freq's influenced by the primary inductance of the TRF and the cap across it and at 500kHz you would have to use a tuned coil to get the desired freq.

This is a very simple inverter using a programmable type oscillator would be a lot better answer, especially if you might need to generate different frequencies for some later projects.

I guess you hvnt ever dealed with such sets! Do you know its difficult and hard?

I do strike as odd. so many enthusiasts out there who like help other with such problem, now all keep silent?

This told you this is a very hard task for you.

At first, you hvnt put forward enough condition for your project. you said you need sine wave, but you hve to tell us how much distortion of the wave is allowed?

The little distortion the difficulty .

you want to get a turns ratio of 1:25, do you know how difficult to wing it at such higher frequency?

you related to the power, that wioo mean it must be a pulse method. how do you think of its control method?

although I can provide you with a reference circuit but I still sugget you told to your boss to buy one on the market.

I know that it´s difficult and hard to obtain a perfect output waveform.

You advised me that I have to buy the circuit. Before buying it I want to try to do it. Buy it´s my last chance.

Thank you for your advise.

So, your actually talking about "Electro stimulation"

There are a few websites that deal with this, and some (tho very entertaining) may have devices that would be along the lines, suitable for what your looking for ;o)

JaiAlai,

having read all your questions and explanations concerning your project, gathering on your electronic experiences from that, I can only strictly recommend you to stop this kind of development. Perhaps I`m wrong - then pls. forgive me, but here are some thoughts about that based on 40 years electronic development experience:

1200V and 100W means body currents of about 80 mA. 1200V is "HIGH VOLTAGE" and even some less milliamps can lead to death.

Such experiments should be in the hands of specialised experts for medical equipment who know that such kind of development requires manyears of experience, not only with a soldering iron but, and predominantly, in terms of medical security aspects, technical approvals, EMI-approvals and clinical tests.

Don`t undervalue the technical problems and, quite more important, the juristic aspects - if you develop a home-brew audio amplifier and it misfunctions, you have lost some electric parts worth some EUROS, if your medical equipment fails, you`ve lost your patient and I would not like to be in his (or your) shoes.

Regards Uwe